The present invention relates to a network system in which a plurality of internetwork apparatuses such as routers each connecting networks at the network layer level are used to connect a plurality of networks to one another, and more particularly to a network system having the redundant configuration in which a current internetwork apparatus is changed to a standby internetwork apparatus upon failure of the current internetwork apparatus.
In the system in which a plurality of networks are connected by a single router, when a failure occurs in the router, operation of the whole of the system is stopped, so that the whole of the system fails to be operated normally. On the other hand, there is a system having the redundant configuration using two routers one serving as a current router and the other serving as a standby router and in which in the normal state of the current router the system is operated by the current router and when a failure occurs in the current router the system is operated by the standby router instead of the current router to thereby attain a high reliability system.
Such a system having the redundant configuration is disclosed in, for example, JP-A-6-131208. In this system, when a failure occurs in a current network apparatus, operation of the whole of the current network apparatus is stopped and the current network apparatus is changed to a standby network apparatus to thereby realize the system having the redundant configuration.
Further, in a network system, various management information such as statistical information, set information and the like is exchanged among nodes in the network and is managed as a management information base (MIB).
In the system having the redundant configuration in the prior art, when a failure occurs in a port of the current network apparatus or in a connection portion between the port and the network, operation of the whole of the current apparatus is stopped. That is, operation of not only the port in which the failure occurs but also other all normal ports in the current network apparatus is stopped and operation is changed from the current network apparatus to the standby network apparatus. Accordingly, even in communication between the networks connected to the normal ports of the current network apparatus, it is necessary to disconnect the normal ports and to change the communication routes to the standby network apparatus.
Further, in this case, when data is transmitted from a terminal of the network, the relay port of the data is changed and accordingly an address thereof must be also changed.
In addition, even in transmission and reception of management information managed by using management information base (MIB), it is necessary to transmit and receive the management information in consideration of the physical configuration of the network apparatus in response to change of the route.
More particularly, in the system for managing networks on the basis of MIB information, that is, information based on the management information base (MIB) for each network, when a failure occurs in a certain port of a current apparatus, operation of the current apparatus is stopped completely and is changed from the current apparatus to the standby apparatus. Accordingly, MIB information of all ports (that is, all networks) of the current apparatus cannot be obtained. Thus, management of the networks cannot be performed on the basis of the MIB information available before occurrence of the failure.
It is an object of the present invention to provide a network system and a failure restoration method of the network system which is adapted to be solve the drawbacks in the prior art.
It is another object of the present invention to provide a network system and a failure restoration method of the network system in which when a failure occurs in a port of a current apparatus, data communication can be made without change of other normal ports of the current apparatus.
It is a further object of the present invention to provide a network system and a failure restoration method of the network system in which even a terminal having no dynamic routing function and address resolution protocol (ARP) function for conversion of address, that is, having only the routing function can make communication without considering change of a route due to occurrence of a failure.
It is another object of the present invention to provide a network system capable of collecting management information base (MIB) information without considering change of a route due to occurrence of a failure.
In order to achieve the above objects, according to an aspect of the present invention, there is provided a network system which includes a plurality of networks, a first internetwork apparatus including a plurality of first ports each connected to the plurality of networks, a second internetwork apparatus including a plurality of second ports each connected to the plurality of networks, and a data transmission unit connected to the first and second internetwork apparatuses to transmit data mutually between the first and second internetwork apparatuses, wherein the first internetwork apparatus includes a first transmitting and receiving unit connected to the plurality of first ports and the data transmission to transmit and receive data mutually among the plurality of first ports and the data transmission unit, a failure detection unit for detecting whether a failure occurs in any of one of the plurality of first ports and a route between one of the plurality of first ports and one of the plurality of networks connected thereto and, when occurrence of a failure is detected, issuing notification indicating a failure occurrence portion to transmit the notification through the first transmitting and receiving unit to the data transmission unit, and a first port control unit for causing each of the plurality of first ports to be able to transmit and receive data to and from the one of the plurality of networks in a normal state and causing the one of the plurality of first ports not to be able to transmit and receive data to and from the one of the plurality of networks in response to detection of occurrence of failure by the failure detection unit, and the second internetwork apparatus includes a second transmitting and receiving unit connected to the plurality of second ports and the data transmission unit for transmitting and receiving data mutually among the plurality of second ports and the data transmission unit and for receiving the notification transmitted through the data transmission unit to produce the notification, and a second port control unit for causing the plurality of second ports not to be able to receive data from the plurality of networks and to be able to transmit data to the plurality of networks in the normal state and responsive to the notification from the second transmitting and receiving unit to cause one of the plurality of second ports connected through the one of the plurality of networks to the one of the plurality of first ports to be able to transmit and receive data to and from the one of the plurality of networks.
According to one example of the present invention, the first port control unit sets an address of each of the plurality of first ports to a common address common to the port and a corresponding one of the plurality of second ports connected through a corresponding one of the plurality of networks in the normal state, the common address being different for each of the plurality of first ports, and the second port control unit sets an address of each of the plurality of second ports to an address inherent to the port in the normal state.
According to one example of the present invention, the first port control unit includes a unit for disconnecting the one of the plurality of first ports from the one of the plurality of networks in response to detection of occurrent of failure by the failure detection unit, and the second port control unit changes an address of the one of the plurality of second ports from the inherent address to the common address set in the one of the plurality of first ports in response to the notification from the second transmitting and receiving unit.
As described above, even if a failure occurs in any portion of a port in the current internetwork apparatus (router), the routing module including the port and a route between the port and the pertinent networks, only the port corresponding to the portion where the failure occurs is stopped or electrically disconnected from the pertinent network and the normal port in the current system is operated as it is. Further, the address of a port of the standby system corresponding to the failed port where the failure occurs is changed to an address of the failed port, that is, the port of the standby system is caused to be able to perform transmission and reception. Thus, change of the transmission route of packet data can be minimized and communication between the networks can be continued. In other words, even if a failure occurs in a portion of the current router, it is not necessary to change the whole current system to the standby system as in the prior art, the route corresponding to only the failed portion is changed and the routes in the normal portions of the current system are not required to be changed.
Further, the relay address of the packet data from a terminal is not required to be changed and the route is automatically changed so that the packet data is transmitted to the terminal of the destination.
In other words, since the relay address of the packet data from the terminal is not required to be changed and the packet data from the terminal is automatically transmitted to the destination terminal, the terminal can perform communication without considering change of the route for the communication.
As described above, according to the present invention, the reliable and inexpensive LAN system can be constructed by realization of duplication at a unit of port. Further, the terminal having neither dynamic routing function nor ARP function can communicate without considering change of a route. In addition, management of the network such as collection of statistical information and the like can be made without considering the redundant configuration.
In order to achieve the above objects, according to another aspect of the present invention, there is provided a network system further comprising a management terminal connected to the one of the plurality of networks for managing management information base (MIB) information of each of the plurality of first and second ports, wherein the first internetwork apparatus further comprises a first MIB information memory unit for storing the management information base information for each of the plurality of first ports, and a first MIB information control unit connected to the first MIB information memory unit and the first transmitting and receiving unit, and the second internetwork apparatus further comprises a second MIB information memory unit for storing management information base (MIB) information for each of the plurality of second ports, and a second MIB information control unit connected to the second MIB information memory unit and the second transmitting and receiving unit, the management terminal sending a collection request of management information base information of the one of the plurality of first ports through the one of the plurality of networks, the one of the plurality of second ports and the second transmitting and receiving unit to the second MIB information control unit after the one of the plurality of second ports has been caused to be able to transmit and receive data to and from the one of the plurality of networks, the second MIB information control unit being responsive to the collection request of the management information base information from the management terminal to read out management information base information from the second MIB information memory unit after the one of the plurality of second ports has been caused to be able to transmit and receive data, and sending a collection request of management information base information available before occurrence of failure of the one of the plurality of first ports through the second transmitting and receiving unit, the data transmission unit and the first transmitting and receiving unit to the first MIB information control unit, the first MIB information control unit being responsive to the collection request of the management information base information from the second MIB information control unit to read out management information base information available before occurrence of failure of the one of the plurality of first ports from the first MIB information memory unit and transmitting the management information base information through the first transmitting and receiving unit, the data transmission unit and the second transmitting and receiving unit to the second MIB information control unit, the second MIB information control unit calculating a sum total of management information base information read out from the first and. second MIB information memory units to send the sum total through the second transmitting and receiving unit, the one of the plurality of second ports and the one of the plurality of networks to the management terminal.
As described above, when the collection request of the statistical information (MIB) is issued from the management terminal of the network to the system having the redundant configuration, the internetwork apparatus operating as the current system reads out the statistical information of the internetwork apparatus serving as the standby system and the statistical information of the whole redundant configuration system is collectively returned to the management terminal. Thus,. the management terminal can manage the networks without considering the physical redundant configuration.
As described above, even if a failure occurs in a port of the current system, operation of only the failed port (or all ports included in the routing module of the failed port) is stopped and other normal ports of the current system are continued to be operated, so that the port of the standby system corresponding to the failed port (or the stopped port) is operated. Accordingly, the past management information base (MIB) concerning the stopped port of the current system can be obtained. Hence, the networks can be managed on the basis of the management information base (MIB) information available before occurrence of the failure.
Further, since the address of the port where the failure occurs is used as the address of the port of the standby system operated instead, the management terminal can obtain the management information base (MIB) information of the port of the current system of which operation is automatically stopped without changing the address of the source requiring the MIB information regardless of change of the route.